BlackRock to Cut Around 250 Jobs as CEO Larry Fink Pushes Strategic Shift 
HSBC Expands UAE Presence With New Asset Management Business and Onshore Funds 
FTC Blocks Edwards Lifesciences’ JenaValve Acquisition in Major Antitrust Ruling 
Viking Therapeutics Sees Growing Strategic Interest in $150 Billion Weight-Loss Drug Market 
Trump Pushes Tech Giants to Absorb AI Data Center Power Costs, Citing Microsoft Changes 
Lynas CEO Amanda Lacaze to Retire After 12 Years as Rare Earths Demand Grows 
UBS Upgrades L’Oréal to Buy, Sees Strong Sales Momentum and 20% Upside 
Supreme Court to Hear Cisco Appeal on Alien Tort Statute and Human Rights Liability 
China’s AI Sector Pushes to Close U.S. Tech Gap Amid Chipmaking Challenges 
Trump Weighs Blocking Exxon Investment as Venezuela Deemed “Uninvestable” 
Chevron Sees Path to Boost Venezuela Oil Output by 50% After Trump Administration Talks 
Walmart to Join Nasdaq-100 Index as It Replaces AstraZeneca Following Exchange Move 
Elon Musk Says X Will Open-Source Its Algorithm Amid EU Scrutiny 
Boeing 737 MAX 10 Advances in FAA Testing as Certification Delays Continue 
Merck Raises Growth Outlook, Targets $70 Billion Revenue From New Drugs by Mid-2030s 
Trump Considers Starlink to Restore Internet Access in Iran Amid Protests 
PALO ALTO, Calif., March 23, 2017 -- Sighthound, Inc. today announced the release of a deeply learned license plate detection and recognition API in its Cloud service and as part of its Sighthound Sentry software. The software scans images or videos and uses Sighthound’s proprietary deep computer vision algorithms to detect and read license plates as well as recognize the country/state of origin. In addition, it also provides confidence scores for each of the above to help with custom development.
The license plate detector can find license plates in a much broader range of real world deployments than traditional single lane focus cameras, and the recognition models are not only robust to license plates from many different parts of the world but also to natural variations in conditions. The software is designed to return the state for US license plates and country for certain EU countries. Demonstrations of the software running on an nVidia TX1 edge device and in the Sighthound Cloud are available at www.sighthound.com.
The deeply learned system is significantly more accurate than any open source or commercial alternatives to which Sighthound had access. Benchmarks are available in a white paper here: https://www.sighthound.com/technology/. The technology draws upon Sighthound’s experience in building fast, deep Convolutional Neural Networks (CNNs) to perform complex tasks accurately with much more speed and much less data than in the past.
License plate recognition joins Sighthound’s vehicle make, model and color recognition in a new Vehicle Product Suite that is available to companies in fields such as law enforcement, traffic control, toll monitoring, and motor vehicle transportation.
Sighthound makes it easy for developers to get started by offering free tutorials, sample code, and documentation to help integrate its APIs into their own applications.
Developers can use the free Sighthound Cloud API at https://www.sighthound.com/products/cloud. Paid accounts (Basic, PRO, and Enterprise) have access to production servers for commercial use and begin at $49 per month.
About Sighthound
Sighthound is the easiest way to add state-of-the-art computer vision software to your products and services. www.sighthound.com
Media Contacts: Sighthound: Rainer Flor, Sighthound, [email protected]
